CN115187527A - Separation and identification method for multi-source mixed ultrahigh frequency partial discharge spectrum - Google Patents
Separation and identification method for multi-source mixed ultrahigh frequency partial discharge spectrum Download PDFInfo
- Publication number
- CN115187527A CN115187527A CN202210721508.5A CN202210721508A CN115187527A CN 115187527 A CN115187527 A CN 115187527A CN 202210721508 A CN202210721508 A CN 202210721508A CN 115187527 A CN115187527 A CN 115187527A
- Authority
- CN
- China
- Prior art keywords
- partial discharge
- map
- prpd
- discharge
- detection
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/12—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing
- G01R31/1227—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing of components, parts or materials
- G01R31/1254—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing of components, parts or materials of gas-insulated power appliances or vacuum gaps
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
- G06V10/763—Non-hierarchical techniques, e.g. based on statistics of modelling distributions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/06—Recognition of objects for industrial automation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Probability & Statistics with Applications (AREA)
- Testing Relating To Insulation (AREA)
Abstract
The invention discloses a separation and identification method of a multi-source mixed ultrahigh frequency partial discharge map, which comprises the following steps: acquiring a PRPD map of the ultrahigh frequency partial discharge signal, carrying out standardization processing, constructing and labeling a partial discharge sample database, and constructing a PRPD map data set of the partial discharge signal and a noise signal; determining a deep learning network as a detection model, and inputting parameters of the detection model; dividing a PRPD atlas data set, inputting the PRPD atlas data set into a deep learning network detection model for training, adjusting a detection model and training parameters, reserving and evaluating the trained model, and selecting the model with the optimal effect as a PRPD atlas detection model; and (4) using the selected detection model to analyze and confirm the detection targets one by one on the atlas to be identified, and keeping the results of final detection and multi-source separation. The invention solves the problem that when an image recognition algorithm is used for carrying out discharge type recognition on a GIS ultrahigh frequency PRPD map, a plurality of partial discharge signals cannot be detected respectively when actually existing.
Description
Technical Field
The invention belongs to the field of partial discharge signal identification, and particularly relates to a separation and identification method of a multi-source mixed ultrahigh frequency partial discharge map based on a digital image target detection and pairing technology.
Background
Since the practical use of Gas Insulated Switchgear (GIS) equipment in the 60 th 20 th century, the equipment has been widely used not only in the high-voltage and ultrahigh-voltage fields but also in the ultra-high-voltage field. However, in a complex working environment or during the manufacturing and installation process, some internal defects and safety hazards such as poor conductor contact, conductive particles, metal tips, air gaps of insulating parts and the like are inevitably generated, so that various types of partial discharge are caused, and further, GIS insulation failure and power system accidents are caused. Therefore, ultrahigh frequency partial discharge detection is regularly carried out on the GIS equipment in operation, partial discharge inside the GIS is analyzed and diagnosed according to the detected partial discharge PRPD map, and hidden danger of defects inside the GIS is timely discovered and eliminated, so that the ultrahigh frequency partial discharge detection method becomes an important component of GIS equipment operation and maintenance.
The type identification of the partial discharge is an important link of GIS partial discharge detection, and is mainly based on a partial discharge PRPD map. At present, for GIS partial discharge type identification, besides manual identification by means of inspection personnel according to experience, many research institutions research and develop artificial intelligent identification methods based on image identification technology to realize automation of partial discharge identification. Such methods have the following problems: and only one partial discharge type is output as a final identification result aiming at the fact that one piece of partial discharge map data can be identified only according to the overall characteristics of the map image. In actual conditions, frequently occurring conditions such as coexistence of multiple partial discharge signals inside a GIS, coexistence of internal partial discharge signals and external noise signals, etc., a PRPD map is presented as a mixed superimposed image of multiple partial discharge signal maps, which causes map 'four-unlike', resulting in recognition errors. Even if the type of a certain signal source is correctly identified, other coexisting signal sources must be missed.
Disclosure of Invention
The invention aims to provide a method for separating and identifying a multi-source mixed ultrahigh frequency partial discharge map, which is used for solving the problem that when the type of a GIS ultrahigh frequency PRPD map is identified by the conventional method, the map with a plurality of coexisting signal sources cannot be detected one by one and each signal source cannot be identified.
A separation and identification method for a multi-source hybrid ultrahigh frequency partial discharge map comprises the following steps:
the method comprises the following steps: acquiring a PRPD map by using an ultrahigh frequency partial discharge detection device, processing and converting the PRPD map into a standard gray image, and constructing a partial discharge map sample database;
step two: marking a sample database, selecting the map areas corresponding to each signal source in each sample image one by one, and marking the corresponding signal types to form a PRPD map data set;
step three: determining a deep learning network as a detection model, and inputting parameters of the detection model;
step four: dividing a PRPD atlas data set, inputting the PRPD atlas data set into a deep learning network detection model for training, adjusting the detection model and training parameters, and keeping the trained model;
step five: evaluating the trained model, and selecting the model with the optimal effect as a PRPD map detection model of the partial discharge signal;
step six: detecting discharge signals of the to-be-detected map, and detecting various targets of partial discharge and noise interference in the corresponding gray level map;
step seven: and analyzing and confirming the detection targets one by one, and reserving the detection targets meeting the standard as the final detection and multi-source separation results.
Preferably, in the first step, a PRPD pattern is obtained by using an ultrahigh frequency partial discharge detection apparatus, and the PRPD pattern includes: independent partial discharge signals and interference noise signals, and a plurality of signals mixing partial discharge and interference noise;
and converting the PRPD map into a gray map, wherein the gray value of each pixel is that the pulse number at the position is normalized to be in a range of [0,255], namely the maximum value of the pulse number is normalized to be 255, and other pulse numbers are normalized to be an integer in a range of [0,254] according to the same proportion.
Preferably, in the second step, the labeling of the partial discharge sample data is performed by an image labeling tool "labelme", wherein: for each map in the partial discharge sample database, respectively marking the region of each partial discharge signal and each noise signal in the PRPD map and the respective signal type by adopting a rectangular frame; the suspension discharge, solid insulation discharge and particle discharge signals are gathered in a PRPD map in two clusters with the phase deviation of 180 degrees and are respectively marked by adopting 2 rectangular frames;
the PRPD profile dataset comprises: 5 targets to be detected, namely suspension discharge, solid insulation discharge, particle discharge, point discharge and interference noise.
Preferably, in the third step, the deep learning detection model is based on the YOLOv3 detection algorithm.
Preferably, in the third step, the darknet-53 is adopted as the backbone network of the YOLOv3 detection algorithm.
Preferably, in the third step, the deep learning network detection model includes feature maps of 13 × 13, 26 × 26, and 52 × 52 pixels, and the size used by the feature maps is an average size of 9 types of prior frames clustered by a K-means algorithm.
Preferably, in step four, the PRPD profile dataset comprises: training set, verification set and test set, wherein: the proportion of the training set, the verification set and the test set is 6:2:2.
preferably, in the fourth step, parameters of batch (the number of pictures input to the detection model for training each time) and the initial learning rate are set, and the learning rate is dynamically adjusted according to the change of the loss function;
the conditions for storing the trained model are as follows: the value of the loss function fluctuates around a small range and does not decrease any more.
Preferably, in the fifth step, the model with the optimal effect is as follows: the highest mAP trained model on the test set.
Preferably, in the seventh step, analyzing and confirming the detection targets one by one specifically includes: for the targets to be detected of suspension discharge, solid insulation discharge and particle discharge, in the areas which are 180 degrees away from the phase and within 10dB of amplitude difference, if similar targets exist, the targets are matched and then used as the same detected partial discharge signal; if no homogeneous object exists, the object is discarded.
The invention has the following technical effects:
1. the problem that when an image recognition algorithm is used for carrying out discharge type recognition on a GIS ultrahigh frequency PRPD map, a plurality of partial discharge signals cannot be detected respectively when actually existing is solved;
2. the multi-source separation can be accurately carried out on the multi-partial discharge mixed signal spectrum, and the type identification can be respectively carried out on the separated signals;
3. the traditional GIS discharge signal detection mode is changed, automatic detection and multi-source separation are achieved, parameters of an initial candidate box are improved by adopting a K-means clustering algorithm, the identification speed is effectively increased, mAP on a test set reaches more than 95%, and the GIS discharge signal can be monitored in real time more accurately and more efficiently.
Drawings
FIG. 1 is a schematic flow diagram of the present invention;
FIG. 2 is a schematic diagram of PRPD map normalization and data set annotation in accordance with the present invention;
in the figure: (a) - (e) are gray scale graphic representations after PRPD atlas standardization, (f) - (j) are data set labeling representations;
FIG. 3 is a schematic diagram of a network structure of a YOLOv3 target detection algorithm of the present invention;
FIG. 4 is a schematic view of the detection process of the to-be-detected atlas of the present invention.
Detailed Description
The technical solution of the present invention is further described below with reference to the following embodiments and the accompanying drawings.
The PRPD map, namely a Phase Resolved Partial Discharge map (Phase Resolved Partial Discharge), is a statistical map formed by collecting a plurality of Partial Discharge ultrahigh-frequency pulses within a period of time by a Partial Discharge detector, and simultaneously collecting and recording the power frequency Phase (0-360 DEG) of a power grid when each pulse is collected. The atlas is a scattered two-dimensional coordinate plane diagram, the abscissa is the power frequency phase (0-360 degrees) of the power grid of the pulse, the ordinate is the power of the pulse (generally ranging from-80 dBm to 0 dBm), and the value of each scattered point is the number of pulses appearing in the acquisition time period at the coordinate.
The embodiment provides a separation and identification method of a multi-source hybrid ultrahigh frequency partial discharge map, which comprises the following steps:
the method comprises the following steps: acquiring a PRPD map by using an ultrahigh frequency partial discharge detection device, processing and converting the PRPD map into a standard gray image, and constructing a partial discharge map sample database;
step two: marking a sample database, performing frame selection on the map areas corresponding to each signal source in each sample image one by one, and marking the corresponding signal types to form a PRPD map data set;
step three: determining a deep learning network as a detection model, and inputting parameters of the detection model;
step four: dividing a PRPD atlas data set, inputting the PRPD atlas data set into a deep learning network detection model for training, adjusting the detection model and training parameters, and keeping the trained model;
step five: evaluating the trained model, and selecting the model with the optimal effect as a PRPD map detection model of the partial discharge signal;
step six: detecting discharge signals of the to-be-detected map, and detecting various targets of partial discharge and noise interference in the corresponding gray level map;
step seven: and analyzing and confirming the detection targets one by one, and reserving the detection targets meeting the standard as the final detection and multi-source separation results.
In a further implementation manner of this embodiment, in the first step, a PRPD map is obtained by using an ultrahigh frequency partial discharge detection apparatus, where the PRPD map includes: four kinds of independent partial discharge signal maps, various kinds of external interference signal maps and various kinds of partial discharge and external interference mixed signal maps;
converting the PRPD map into a gray map, and obtaining five thousand gray maps corresponding to the PRPD map in total, wherein ten thousand gray maps of the four independent partial discharge signals and ten thousand gray maps of the mixed partial discharge signals are obtained; the abscissa and the ordinate are respectively quantized to 416 parts, the gray value of each pixel is the range of normalizing the pulse number at the position to be [0,255], namely the maximum value of the pulse number is normalized to be 255, and other pulse numbers are normalized to be an integer in the range of [0,254] according to the same proportion.
In a further implementation manner of this embodiment, in step two, an image annotation tool "labelme" is used to perform annotation on partial discharge sample data, and an xml file including annotation information is generated, where: for each map in the partial discharge sample database, respectively marking the region of each partial discharge signal and each noise signal in the PRPD map and the respective signal type by adopting a rectangular frame; two clusters of signals of suspension discharge, solid insulation discharge and particle discharge with phase deviation of 180 degrees are gathered in a PRPD map, and are respectively marked by adopting 2 rectangular frames;
the PRPD profile dataset includes: 5 targets to be detected, namely suspension discharge, solid insulation discharge, particle discharge, point discharge and interference noise.
In a further implementation manner of this embodiment, in step three, the deep learning detection model is based on the YOLOv3 detection algorithm, the size parameter of the input picture is 416 × 416 pixels, yolo is a target detection algorithm, the task of target detection is to find out an object from the picture and give its category and position, it applies a single Convolutional Neural Network (CNN) to the whole image, divides the image into meshes, and predicts the class probability and bounding box of each mesh.
In a further implementation manner of this embodiment, in step three, the darknet-53 is used as the backbone network of the YOLOv3 detection algorithm, the darknet-53 includes 53 layers of convolutions, the first 52 layers of convolutions are set as the main network, and the last layer is the full link layer.
In a further implementation manner of this embodiment, in step three, the deep learning network detection model includes feature maps of 13 × 13, 26 × 26, and 52 × 52 pixels, and the size used by the feature maps is an average size of 9 types of prior frames clustered by the K-means algorithm.
The number of the prior frames is 13 × 3+26 × 3+52 × 3=10647, each prediction is a 10-dimensional vector (4 +1+ 5=10), and the 10-dimensional vector includes: frame coordinates (4 values), frame confidence (1 value), probability of object class (4 discharge signals +1 noise signal);
the final output depth of each feature map is: 3 (+1 + 5) =30;
the loss function includes: center coordinate error, width and height coordinate error, confidence error, and category error.
In a further embodiment of this example, in step four, the PRPD profile dataset comprises: training set, verification set and test set, wherein: the proportion of the training set, the verification set and the test set is 6:2:2, 30000, 10000 and 10000 respectively.
In the fourth step, batch is set to 64, which means that network parameters are updated by performing forward propagation once per 64 training samples, the total number of iterations is set to 400000, the initial learning rate is 0.001, the learning rate is dynamically adjusted according to the change of the loss function, and when the iteration is 200000, the learning rate is attenuated by ten times, and when the iteration is 300000, the learning rate is attenuated by ten times on the basis of the previous learning rate;
the conditions for storing the trained model are as follows: the value of the loss function fluctuates around a small range and does not decrease any more, and the number of saved models is 5.
In a further implementation manner of this embodiment, in step five, the five trained models are respectively tested on the test set, an mAP (average accuracy of the five types of targets) is calculated, and a model with the highest mAP is selected as a final detection model.
In a further implementation manner of this embodiment, in step seven, analyzing and confirming the detection targets one by one specifically includes: only two targets are detected for the target to be detected of the point discharge signal and the interference noise signal, and the detection result is reserved; for the targets to be detected of suspension discharge, solid insulation discharge and particle discharge, in the area (corresponding to 208 pixels for transverse translation and 52 pixels for longitudinal difference on a gray scale map) which is 180 degrees away from the phase and within 10dB of the amplitude, if the same type of targets exist, the targets are paired and then used as the same detected partial discharge signal; if no homogeneous object exists, the object is discarded.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A separation and identification method for a multi-source hybrid ultrahigh frequency partial discharge map is characterized by comprising the following steps:
the method comprises the following steps: acquiring a PRPD map of a GIS partial discharge signal, and carrying out standardization processing on the PRPD map to construct a partial discharge sample database;
step two: labeling a partial discharge sample database, and constructing a PRPD map data set of a discharge signal and a noise signal;
step three: determining a deep learning network as a detection model, and inputting parameters of the detection model;
step four: dividing a PRPD atlas data set, inputting the PRPD atlas data set into a deep learning network detection model for training, adjusting the detection model and training parameters, and keeping the trained model;
step five: evaluating the trained model, and selecting the model with the optimal effect as a PRPD map detection model of the partial discharge signal;
step six: detecting discharge signals of the to-be-detected map, and detecting various targets of partial discharge and noise interference in the corresponding gray level map;
step seven: and analyzing and confirming the detection targets one by one, and reserving the detection targets meeting the standard as the final detection and multi-source separation results.
2. The method for separating and identifying the multi-source hybrid ultrahigh frequency partial discharge map according to claim 1, wherein in the first step, the ultrahigh frequency partial discharge detection device is used to obtain a PRPD map, and the PRPD map comprises: independent partial discharge signals and interference noise signals, and a plurality of signals mixing partial discharge and interference noise;
and converting the PRPD map into a gray scale map, wherein the range of the gray scale value normalized by the pulse number of each pixel at the position is [0,255].
3. The method for separating and identifying the multi-source hybrid ultrahigh frequency partial discharge atlas of claim 1, wherein in the second step, labeling of the partial discharge sample database is performed by an image labeling tool, wherein: for each map in the partial discharge sample database, respectively marking the positions of the partial discharge signal and the interference noise signal in the PRPD map and the signal type by adopting a rectangular frame; the suspension discharge, solid insulation discharge and particle discharge signals are gathered in two clusters with phase deviation of 180 degrees in a PRPD map and are respectively marked by adopting 2 rectangular frames;
the PRPD profile dataset comprises: suspension discharge, solid insulation discharge, particle discharge, point discharge and interference noise.
4. The method for separately identifying the multi-source hybrid ultrahigh frequency partial discharge spectrum according to claim 1, wherein in the third step, the deep learning detection model is based on a YOLOv3 detection algorithm.
5. The method for separating and identifying the multi-source hybrid ultrahigh frequency partial discharge spectrum according to claim 4, wherein in the third step, darknet-53 is adopted as a backbone network of a YOLOv3 detection algorithm.
6. The method for separating and identifying the multi-source hybrid ultrahigh frequency partial discharge spectrum according to claim 5, wherein in the third step, the deep learning network detection model comprises a feature map of 3 types of pixels, and the feature map uses an average size of 9 types of prior frames clustered by a K-means algorithm.
7. The method for separating and identifying the multi-source hybrid ultrahigh frequency partial discharge spectrum according to claim 1, wherein in the fourth step, the PRPD spectrum data set comprises: training set, verification set and test set, wherein: the proportion of the training set, the verification set and the test set is 6:2:2.
8. the method for separating and identifying the multi-source hybrid ultrahigh frequency partial discharge spectrum according to claim 1, wherein in the fourth step, the number of pictures and the initial learning rate for training by inputting the detection model each time are set, and the learning rate is dynamically adjusted according to the change of the loss function;
the conditions for storing the trained model are as follows: the value of the loss function fluctuates around a small range and does not decrease any more.
9. The method for separating and identifying the multi-source hybrid ultrahigh frequency partial discharge spectrum according to claim 1, wherein in the step five, the model with the optimal effect is as follows: the trained model with the highest mAP on the test set.
10. The method for separating and identifying the multi-source hybrid ultrahigh frequency partial discharge spectrum according to claim 1, wherein in the seventh step, the step of analyzing and confirming the detection targets one by one specifically comprises the following steps: for the targets to be detected of suspension discharge, solid insulation discharge and particle discharge, in the areas which are 180 degrees away from the phase and within 10dB of amplitude difference, if similar targets exist, the targets are matched and then used as the same detected partial discharge signal; if no homogeneous object exists, the object is discarded.
Priority Applications (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210721508.5A CN115187527B (en) | 2022-06-27 | 2022-06-27 | Separation and identification method for multi-source mixed ultrahigh frequency partial discharge spectrum |
| PCT/CN2023/101087 WO2023213332A1 (en) | 2022-06-27 | 2023-06-19 | Separation and identification method for multi-source hybrid ultra-high frequency partial discharge diagram |
| CN202380010710.XA CN117242487A (en) | 2022-06-27 | 2023-06-19 | Separation and identification method for multisource mixed ultrahigh frequency partial discharge spectrum |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210721508.5A CN115187527B (en) | 2022-06-27 | 2022-06-27 | Separation and identification method for multi-source mixed ultrahigh frequency partial discharge spectrum |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115187527A true CN115187527A (en) | 2022-10-14 |
| CN115187527B CN115187527B (en) | 2023-04-07 |
Family
ID=83515178
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210721508.5A Active CN115187527B (en) | 2022-06-27 | 2022-06-27 | Separation and identification method for multi-source mixed ultrahigh frequency partial discharge spectrum |
| CN202380010710.XA Pending CN117242487A (en) | 2022-06-27 | 2023-06-19 | Separation and identification method for multisource mixed ultrahigh frequency partial discharge spectrum |
Family Applications After (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202380010710.XA Pending CN117242487A (en) | 2022-06-27 | 2023-06-19 | Separation and identification method for multisource mixed ultrahigh frequency partial discharge spectrum |
Country Status (2)
| Country | Link |
|---|---|
| CN (2) | CN115187527B (en) |
| WO (1) | WO2023213332A1 (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116091501A (en) * | 2023-04-07 | 2023-05-09 | 武汉纺织大学 | Method, device, equipment and medium for identifying partial discharge type of high-voltage electrical equipment |
| WO2023213332A1 (en) * | 2022-06-27 | 2023-11-09 | 上海格鲁布科技有限公司 | Separation and identification method for multi-source hybrid ultra-high frequency partial discharge diagram |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117630611B (en) * | 2024-01-22 | 2024-04-12 | 南京卓煊电力科技有限公司 | Full-bandwidth high-frequency partial discharge PRPD spectrogram capturing and generating method and system |
| CN117607642B (en) * | 2024-01-23 | 2024-03-26 | 中铁电气化铁路运营管理有限公司 | Rail transit data analysis method and system |
| CN117786521B (en) * | 2024-02-23 | 2024-04-19 | 兰州交通大学 | Signal classification method based on intelligent frequency spectrum sensing |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104198899A (en) * | 2014-08-04 | 2014-12-10 | 西安交通大学 | Partial discharge type identifying method for transformer under multi-discharge source situation |
| CN105807190A (en) * | 2014-12-29 | 2016-07-27 | 国家电网公司 | GIS partial discharge ultrahigh frequency live-line detection method |
| CN110794264A (en) * | 2019-10-14 | 2020-02-14 | 浙江浙能技术研究院有限公司 | Generator partial discharge type identification method based on time domain pulse characteristics |
| CN110850244A (en) * | 2019-11-11 | 2020-02-28 | 国网湖南省电力有限公司 | Local discharge defect time domain map diagnosis method, system and medium based on deep learning |
| CN111142001A (en) * | 2020-01-10 | 2020-05-12 | 三峡大学 | Transformer multi-source partial discharge mode identification method based on parallel characteristic domain |
| US20200271714A1 (en) * | 2019-02-22 | 2020-08-27 | Ormazabal Corporate Technology, A.I.E. | Method and system of partial discharge recognition for diagnosing electrical networks |
| CN112149549A (en) * | 2020-09-18 | 2020-12-29 | 国网山东省电力公司泰安供电公司 | GIS partial discharge type identification method based on depth residual error network |
| CN112198399A (en) * | 2020-09-07 | 2021-01-08 | 红相股份有限公司 | Identification method and terminal for multi-source electromagnetic wave signals |
| CN114386499A (en) * | 2021-12-30 | 2022-04-22 | 重庆邮电大学 | Multi-source partial discharge signal data stream clustering separation method based on GIS |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103558529B (en) * | 2013-11-14 | 2016-11-23 | 国家电网公司 | A kind of mode identification method of three-phase cartridge type supertension GIS partial discharge altogether |
| CN106556781A (en) * | 2016-11-10 | 2017-04-05 | 华乘电气科技（上海）股份有限公司 | Shelf depreciation defect image diagnostic method and system based on deep learning |
| KR102389439B1 (en) * | 2019-04-01 | 2022-04-22 | 명지대학교 산학협력단 | RNN based partial discharge detection apparatus and method in gas-insulated switchgear |
| CN114186589A (en) * | 2021-12-08 | 2022-03-15 | 国网上海市电力公司 | Superconducting cable partial discharge mode identification method based on residual error network Resnet50 |
| CN114417926A (en) * | 2022-01-19 | 2022-04-29 | 山东大学 | Power equipment partial discharge pattern recognition method and system based on deep convolution generation countermeasure network |
| CN115187527B (en) * | 2022-06-27 | 2023-04-07 | 上海格鲁布科技有限公司 | Separation and identification method for multi-source mixed ultrahigh frequency partial discharge spectrum |
-
2022
- 2022-06-27 CN CN202210721508.5A patent/CN115187527B/en active Active
-
2023
- 2023-06-19 WO PCT/CN2023/101087 patent/WO2023213332A1/en active Application Filing
- 2023-06-19 CN CN202380010710.XA patent/CN117242487A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104198899A (en) * | 2014-08-04 | 2014-12-10 | 西安交通大学 | Partial discharge type identifying method for transformer under multi-discharge source situation |
| CN105807190A (en) * | 2014-12-29 | 2016-07-27 | 国家电网公司 | GIS partial discharge ultrahigh frequency live-line detection method |
| US20200271714A1 (en) * | 2019-02-22 | 2020-08-27 | Ormazabal Corporate Technology, A.I.E. | Method and system of partial discharge recognition for diagnosing electrical networks |
| CN110794264A (en) * | 2019-10-14 | 2020-02-14 | 浙江浙能技术研究院有限公司 | Generator partial discharge type identification method based on time domain pulse characteristics |
| CN110850244A (en) * | 2019-11-11 | 2020-02-28 | 国网湖南省电力有限公司 | Local discharge defect time domain map diagnosis method, system and medium based on deep learning |
| CN111142001A (en) * | 2020-01-10 | 2020-05-12 | 三峡大学 | Transformer multi-source partial discharge mode identification method based on parallel characteristic domain |
| CN112198399A (en) * | 2020-09-07 | 2021-01-08 | 红相股份有限公司 | Identification method and terminal for multi-source electromagnetic wave signals |
| CN112149549A (en) * | 2020-09-18 | 2020-12-29 | 国网山东省电力公司泰安供电公司 | GIS partial discharge type identification method based on depth residual error network |
| CN114386499A (en) * | 2021-12-30 | 2022-04-22 | 重庆邮电大学 | Multi-source partial discharge signal data stream clustering separation method based on GIS |
Non-Patent Citations (1)
| Title |
|---|
| 史强: ""基于t-SNE与CFSFDP算法的多源局部放电脉冲分类技术"", 《中国电力》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023213332A1 (en) * | 2022-06-27 | 2023-11-09 | 上海格鲁布科技有限公司 | Separation and identification method for multi-source hybrid ultra-high frequency partial discharge diagram |
| CN116091501A (en) * | 2023-04-07 | 2023-05-09 | 武汉纺织大学 | Method, device, equipment and medium for identifying partial discharge type of high-voltage electrical equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2023213332A1 (en) | 2023-11-09 |
| CN115187527B (en) | 2023-04-07 |
| CN117242487A (en) | 2023-12-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115187527B (en) | Separation and identification method for multi-source mixed ultrahigh frequency partial discharge spectrum | |
| US11586913B2 (en) | Power equipment fault detecting and positioning method of artificial intelligence inference fusion | |
| CN111797890A (en) | Method and system for detecting defects of power transmission line equipment | |
| CN112034310A (en) | Partial discharge defect diagnosis method and system for combined electrical appliance | |
| CN113344134B (en) | Low-voltage distribution monitoring terminal data acquisition abnormality detection method and system | |
| CN112949715A (en) | SVM (support vector machine) -based rail transit fault diagnosis method | |
| CN112085071A (en) | Power distribution room equipment fault analysis and pre-judgment method and device based on edge calculation | |
| CN111598942A (en) | Method and system for automatically positioning electric power facility instrument | |
| CN104680185B (en) | Hyperspectral image classification method based on boundary point reclassification | |
| CN111881970A (en) | Intelligent outer broken image identification method based on deep learning | |
| CN112668612A (en) | Partial discharge signal clustering analysis method based on grids | |
| CN114700587B (en) | Missing welding defect real-time detection method and system based on fuzzy inference and edge calculation | |
| CN110660049A (en) | Tire defect detection method based on deep learning | |
| CN116231504A (en) | Remote intelligent inspection method, device and system for booster station | |
| CN117368651B (en) | Comprehensive analysis system and method for faults of power distribution network | |
| CN110672988A (en) | Partial discharge mode identification method based on hierarchical diagnosis | |
| CN116977807A (en) | Multi-sensor fusion-based intelligent monitoring system and method for refrigerator | |
| CN116466408B (en) | Artificial neural network superbedrock identification method based on aeromagnetic data | |
| CN113781436B (en) | High-voltage switch state monitoring method based on camera and angle measurement | |
| CN112818596A (en) | Intelligent interference identification positioning method and system thereof | |
| CN111737993A (en) | Method for extracting health state of equipment from fault defect text of power distribution network equipment | |
| CN115082444B (en) | Copper pipe weld defect detection method and system based on image processing | |
| CN111025100A (en) | Transformer ultrahigh frequency partial discharge signal mode identification method and device | |
| CN113960700B (en) | Objective inspection, statistics and analysis system for regional numerical forecasting result | |
| CN115937079A (en) | YOLO v 3-based rapid detection method for defects of power transmission line |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |